Tuberculosis (TB) remains the single most common infectious disease cause of mortality. We propose to examine the inter- relationships of nutrition, immunology, and epidemiology with respect to TB in Tanzania. Given that TB is so much linked with HIV immunologically, clinically, and epidemiologically, it is essential to examine how these relationships are modified by HIV infection. Published animal and human studies suggest that vitamin deficiency is associated with poor immune response in TB. By modulating immune function, nutritional supplements may be a useful adjunct to anti-TB drugs, and could lead to the development of shorter drug regimens. We propose to enroll 600 patients, half of whom have active TB and are co-infected with HIV, and the other half have TB alone. All patients will be randomized to receive either multivitamins or placebo from the start of their TB therapy, through the 8 months of anti-TB therapy, and for the 12 months following that (i.e. for a total of 20 months). Consenting subjects who present to 2 TB clinics in Dar es Salaam seeking care for respiratory illness will be screened for TB and HIV infection and invited to participate in the study. Patients who are severely debilitated (Karnofsky score of 40 percent or less, basal metabolic index of less than 18 kg/m2, or hemoglobin less than 7 g/dl) will be excluded. All patients will receive standard anti-TB therapy. Follow-up visits will occur every two weeks for the first 2 months and monthly thereafter till the end of the study. The minimum duration of followup is 20 months including 8 months of anti-TB treatment. The endpoints of interest include immune response parameters, bacteriologic cure, and clinical outcomes. Immune parameters (CD4 and CD8 cell counts; ex vivo lymphocyte proliferation; and cytokine production, namely IL-2, INF-g, IL-12, and TNF-alpha will be determined at baseline, 2 months, and 6 monthly thereafter. Bacteriologic cure at 1, 2, and 8 months will be determined by ascertaining conversion of sputum culture to negative. Compliance with therapy will be monitored. Surrogate markers are needed to rapidly evaluate the efficacy of anti-TB treatments or vaccines. Hence, we will examine the utility of the above immune response parameters as surrogate markers for treatment efficacy in TB. Using computer simulation modeling and epidemiological data collected during the follow up in this cohort study, we also propose to assess the effectiveness and cost-effectiveness of different treatment strategies for active TB and the use of INH post-TB treatment in Tanzania.